Abstract PD02-04: Molecular Profiling of Circulating Tumor Cells in Blood of Metastatic Breast Cancer Patients Indicates Therapy Response and Provides Information on Epithelial Mesenchymal Transition and Tumor Stem Cell Metabolism

Abstract

Abstract Background: Stem cell like tumor cells may undergo phenotypic changes, known as epithelial-mesenchymal transition (EMT), for metastasis formation using different signal transduction pathways. Assuming that metastasis requires a dissemination of tumor stem cells or tumor cells showing EMT, it was the purpose of this study to evaluate the expression of the stem cell marker ALDH1 and markers involved in two different pathways for EMT in 137 blood samples of 34 metastatic breast cancer patients during a follow-up of palliative chemo-, antibody — or hormonal therapy. Results were correlated with the presence of circulating tumor cells (CTCs) and response to therapy. Materials and Methods: 2 x 5 ml blood were analyzed for CTCs with the AdnaTest BreastCancer (AdnaGen AG). The recovered c-DNA was additionally multiplex tested for the EMT markers [Twist, Akt2, PI3K (EMT1)], [Jagged, Slug, Snail (EMT2)] and separately for the stem cell marker ALDH1. The analytical sensitivity was determined by the detection of 5 target (IGROV) cells spiked into 5 ml blood of healthy donors using the AdnaTest BreastCancer procedure. The identification of EMT markers was considered positive if at least one marker was detected in the sample. Healthy donor samples without spiked tumor cells were used to determine the specificity of the test. The expression of CD34 was analyzed in a subset of samples to exclude potential interference of normal hematopoietic stem cells. Results: 97% of 30 healthy donor samples investigated were negative for EMT1, 90% for EMT2 and 95% for ALDH1 transcripts, respectively. The spiking experiments revealed 80% recovery of the IGROV cells. In total, CTCs were detected in 35/137 (26%) blood samples. In 13/14 (93%) patients responding to a given therapy, no CTCs were detected or dissapeared in the course of the therapy. In contrast, 10/20 patients (50%) not responding to therapy were detected positive for CTCs. All samples were further examined for EMT1, EMT2 and ALDH1, respectively. A subset of samples was tested for CD34 expression with none of the samples found positive. Interestingly, EMT2 markers were only detected in 2/34 (5%) patients. Thus, the following calculations refer to EMT1. In the CTC (+) group, 59% (20/34) of the samples were positive for at least one of the EMT markers or ALDH1 but also in the CTC (-) group, EMT1 or ALDH1 were found in 34% (31/90) of the samples investigated, respectively. In patients diagnosed as non-responders and found positive for EMT markers and ALDH1 during therapy, persistence or reappearance of CTCs was found in 8/9 cases (88%) whereas in the responder group negative for EMT1 and ALDH1, CTCs persisted in only 10% (1/10) of the cases. Conclusion: In metastatic breast cancer (1) a major proportion of CTCs shows EMT and tumor stem cell characteristics (2) the PI3K/AKT pathway and TWIST as a transcription factor seem to be dominant (3) EMT markers or ALDH1 were also detectable in a group of patients negative for CTCs assuming that such markers could indicate a tumor cell phenotype that is not anymore detectable by epithelial markers and thus, could add additional sensitivity for CTC detection. Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr PD02-04.